Search
160 articles for “scalable algorithms”
-
Accelerating Unsupervised Feature Learning: Parallelized Training of Denoising Autoencoders
Abstract: Unsupervised representation learning has become a cornerstone of contemporary machine learning, enabling algorithms to extract informative features from un-labelled, high-dimensional data. This work investigates the efficacy of stacked denoising autoencoders (SDAEs) trained via parallelized stochastic gradient descent (SGD) as a scalable approach to feature extraction. By strategically leveraging multi-threaded computation, our study systematically examines the trade-offs between increased parallelism, training efficiency, and the preservation of model accuracy. Experiments on the …
Published in Journal of Image Processing & Pattern Recognition Progress · Vol. 12, Issue 3, 2025 · pp. 56–64 Read article
-
KSK Approach: An AI-Driven IoT Based Decision Making System’s Study
Abstract: Internet of Things (IoT) has promised a world of interrelated devices, generating vast amounts of data. Traditionally, IoT systems trusted on preprogrammed procedures and human intervention to process data and make decisions. This approach often struggled to hold the sheer size and density of IoT data, leading to inefficiencies and missed opportunities. However, the true budding of that data lies not simply in its collection, but in its interpretation and …
Published in Current Trends in Signal Processing · Vol. 15, Issue 2, 2025 · pp. 14–25 Read article
-
Data-Driven Predictive Analytics and Decision- Making in FinTech Using MongoDB and High-Throughput Data Pipelines
Abstract: This paper examines the implementation of MongoDB and high-throughput data pipelines within the financial technology (FinTech) sector to drive data-informed predictive analytics and decision-making. The study focuses on the architectural components, scalability, and challenges of integrating NoSQL databases into real-time data ingestion and analytics pipelines. The transformative potential of these technologies in modern financial systems is highlighted through practical use cases such as fraud detection, credit scoring, and personalized financial …
Published in International Journal of Algorithms Design and Analysis Review · Vol. 3, Issue 1, 2025 · pp. 1–15 Read article
-
Leveraging Deep Learning for Accurate Weed Identification
Abstract: Weed control is very important for all types of agricultural businesses. The project here revolves around the application of computer vision techniques and, more concretely, deep learning techniques, for the effective recognition and classification of weeds. The EfficientNetB4 architecture is an appropriate backbone as its scalability and performance optimization is adequate. The modifier used is Adam optimization algorithm which will serve as a pre- processor for the model. Weeds at …
Published in Research & Reviews : Journal of Agricultural Science and Technology · Vol. 14, Issue 2, 2025 · pp. 90–99 Read article
-
PREDICTIVE MAINTENANCE IN SEMICONDUCTOR SYSTEMS: INSIGHTS FROM MACHINE INTELLIGENCE AND DATA-DRIVEN METHODS
Abstract: With the fast-paced development of semiconductor technology comes the need to focus on device reliability, or how long devices will function and the likelihood of devices having operational issues. Predicting failures and avoiding downtime with the implementation of timely, actionable, and data-driven maintenance strategies are essential to insure devices function sustainably within predetermined performance levels. The implementation of predictive maintenance within artificial intelligence and machine learning technologies will provide the …
Published in Journal of Semiconductor Devices and Circuits · Vol. 13, Issue 1, 2026 · pp. 1–9 Read article
-
Recent Advances in Content-based Image Retrieval: Techniques and Applications
Abstract: Content-based image retrieval (CBIR) plays a vital role in computer vision, driven by the increasing need for fast and accurate image retrieval across fields like healthcare, e-commerce, and digital libraries. This study offers a detailed review of CBIR methodologies, charting their progression from traditional feature extraction techniques, such as Local Binary Patterns (LBP), to contemporary deep learning-driven methods. The transformative impact of convolution neural networks (CNNs) is highlighted, emphasizing their …
Published in Journal of Computer Technology & Applications · Vol. 16, Issue 1, 2025 · pp. 67–71 Read article
-
Smart Air Filtration Systems for Cities: A Technological Approach to Reducing Urban Pollution
Abstract: Urban air pollution is considered one of the main ecologically critical issues of the 21st century since it threatens citizens' health through respiratory diseases, pathologies of the cardiovascular system, and even premature death. Regarding an extremely high level of pollution in large cities, it becomes necessary to realize technological solutions which could avoid the negative impact of this phenomenon. Smart air filtration systems have emerged as promising solutions for urban …
Published in International Journal of Environmental Noise and Pollution Control · Vol. 2, Issue 2, 2024 · pp. 27–42 Read article
-
Advanced Airline Operations Control System Using NodeJS
Abstract: This research paper presents the development of an Advanced Airline Operations Control System (AAOCS) using NodeJS, a lightweight and scalable JavaScript runtime environment. Leveraging NodeJS's event-driven architecture and non-blocking I/O model, the system facilitates efficient management of flight operations, crew scheduling, resource allocation, and real-time decision-making in the aviation industry. Key features include real-time decision support tools, a microservices-based architecture for scalability, integration with external data sources and APIs, and …
Published in Journal of Control & Instrumentation · Vol. 15, Issue 1, 2024 · pp. 45–51 Read article
-
Proof-of-Minimum Privacy Leak Consensus Strategy in Blockchain
Abstract: In this study, we propose a novel consensus algorithm to preserve the security and privacy of a transaction. We propose a Proof-of-Minimum Privacy Leak consensus strategy. This means that the competing nodes which participate in the competition to mine the next block should give a proof of minimum privacy leak during its transaction. Only this proof will give highest votes to that node, and it will be elected as the …
Published in E-Commerce for Future & Trends Read article
-
Adaptive Machine Learning Framework for Navigation Control of Autonomous Drones
Abstract: The rise of autonomous drones has expanded UAV applications across sectors like surveillance, delivery, agriculture, and rescue operations. However, traditional navigation systems face limitations in adapting to dynamic environments. This study proposes an AI-driven adaptive navigation framework that leverages real-time sensor data, reinforcement learning, and adaptive control strategies to enhance drone autonomy, scalability, and security. The system processes mission inputs, environmental data (from LiDAR, cameras, GPS, and weather sensors), and …
Published in Journal of Microwave Engineering and Technologies · Vol. 12, Issue 3, 2025 · pp. 1–7 Read article
-
Advanced Computational Models for Predicting Molecular Interactions
Abstract: Understanding molecular interactions is essential for a number of disciplines, including biochemistry, materials science, and medication development. Traditional experimental methods, while accurate, are often time-consuming and expensive. Advanced computational models have emerged as powerful tools to predict molecular interactions efficiently. In order to predict the behavior and interactions of molecules at the atomic and subatomic levels, this paper reviews the most recent developments in computational techniques, such as machine learning …
Published in International Journal of Advance in Molecular Engineering · Vol. 2, Issue 1, 2024 · pp. 8–13 Read article
-
Advancements, Hurdles, and Applications in Quantum Computing
Abstract: Using the ideas of quantum mechanics, quantum computing has become a paradigm shift in computing, enabling computations to be completed tenfold quicker than with traditional computers. This study investigates the current status of quantum computing, looking at the notable advancements, ongoing difficulties, and potential uses that could completely change a range of industries. By utilizing the principles of superposition and entanglement, quantum computers can potentially solve complex problems that classical …
Published in Recent Trends in Parallel Computing · Vol. 11, Issue 2, 2024 · pp. 11–29 Read article
-
AI, Robotics, and the Future of Waste Management: A Systematic Review of Advanced Collection and Sorting Systems
Abstract: The rapid growth of cities and rise in population have made waste management a major concern that calls for innovative and efficient solutions. Conventional waste collecting techniques are dangerous, time-consuming, and frequently ineffective. The development of automated waste management systems powered by cutting-edge technology like robotics, deep learning, artificial intelligence (AI), and the Internet of Things (IoT) is examined in this study. Vision-based systems, convolutional neural networks (CNN) for garbage …
Published in International Journal of Robotics and Automation in Mechanics · Vol. 3, Issue 1, 2025 · pp. 1–6 Read article
-
Feature Selection Using Nature-Inspired Metaheuristics: A Brief Study
Abstract: In recent decades, feature selection (FS) has attracted a lot of attention. The rapid advancement of data and computer science has led to an increase in the number of studies being conducted. The preprocessing method of feature selection plays a significant role in increasing the effectiveness of data mining and analysis across significant real-world applications. To summarize the most recent studies, this research work offers a quick assessment on feature …
Published in Current Trends in Information Technology · Vol. 13, Issue 1, 2023 · pp. 12–21 Read article
-
Development and Experimental Investigation of a Light Weight Hybrid Electric Vehicle Using Lithium-Ion Battery and EDLC Supercapacitor Integration
Abstract: Lightweight hybrid electric vehicles (HEVs) have gained significant attention as an environmentally friendly and efficient mode of transportation. However, their energy storage systems (ESS) often face challenges such as limited battery lifespan, inadequate power delivery during peak demand, and inefficient energy recovery during braking. This research focuses on the development and experimental evaluation of a hybrid energy storage system (HESS) integrating Lithium-Ion Batteries (LIB) and Electric Double-Layer Capacitors (EDLC) for …
Published in Journal of Automobile Engineering and Applications · Vol. 12, Issue 1, 2025 · pp. 26–40 Read article
-
A Conceptual Framework for AI-Integrated Metabolomics in Predictive Health Systems for Resource-Constrained Environments
Abstract: The increasing prevalence of non-communicable diseases (NCDs) continues to place a significant strain on healthcare systems, particularly in low- and middle-income regions where access to early diagnostic infrastructure is limited. Conventional healthcare approaches remain largely reactive, often detecting diseases at advanced stages when treatment effectiveness is reduced. This challenge underscores the need for predictive, cost-effective, and data-driven healthcare solutions. This study presents a conceptual framework that integrates metabolomics with artificial …
Published in Emerging Trends in Metabolites · Vol. 3, Issue 2, 2026 Read article
-
Optimization of Pesticide Requirement Calculations for IoT-Operated Hexacopter Delivery Systems
Abstract: The integration of Internet of Things (IoT) technology into precision agriculture has transformed pesticide application strategies, enabling resource-efficient and environmentally sustainable practices. This study presents a computational methodology for optimizing pesticide requirements in an IoT-operated hexacopter system, designed for dynamic, data-driven pesticide delivery. Leveraging a fusion of real-time telemetric data from onboard LiDAR, multispectral imaging sensors, and environmental monitoring modules, the system employs predictive analytics and edge computing to calculate …
Published in International Journal on Drones · Vol. 2, Issue 1, 2026 · pp. 08–14 Read article
-
Mario Ai Model Using Gaming Reinforcement Learning
Abstract: It is essential for research on computational and/or artificial intelligence (CI/AI) applied to games to have relevant games to apply AI algorithms to. This is pertinent. It doesn't matter if one is studying how to use CI/AI techniques to test and improve AI (e.g., games provide challenging yet scalable problems which engage many central aspects of human cognitive capacity) or how to use CI/AI techniques to improve games (e.g., player …
Published in Journal of Instrumentation Technology & Innovations · Vol. 14, Issue 2, 2024 · pp. 1–6 Read article
-
A Low-Cost Multi-Sensor IoT System for Real-Time Segregation of Polymer Waste
Abstract: Segregation of solid waste is a critical aspect of waste management, especially in settings where technical and financial constraints limit the adoption of sophisticated technologies. The proposed low cost, sensor-driven smart waste sorting system combines a variety of sensing technologies with an integrated decision-making system. The system employs an inductive sensor, moisture sensor and capacitive sensor to measure the physical properties of waste items, allowing segregation into metal, wet and …
Published in Journal of Polymer & Composites · Vol. 14, Issue 3, 2026 · pp. 90–`107 Read article
-
Real-Time Object Detection and Tracking in Traffic Surveillance: Implementing Algorithms That Can Process Video Streams for Immediate Traffic Monitoring
Abstract: The rapid growth in urban development and traffic congestion calls for adopting high standards of traffic surveillance systems for monitoring. This paper reviews the current advancement and future trends of real-time object detection and tracking technology and its implications for traffic surveillance. Conventional approaches to traffic monitoring can provide more or less accurate data, but they are not easily scalable and cannot cope with rapidly changing conditions typical within urban …
Published in International Journal of Algorithms Design and Analysis Review · Vol. 3, Issue 1, 2025 · pp. 18–39 Read article